Antibody Microarrays for Biomarker Discovery

The objective of this project is to discover proteins in human serum that can be used as biomarkers for cancer early detection. Our strategy for biomarker discovery utilizes over 1000 full length antibodies and 2000 single chain variable fragment (scFv) antibodies selected against specific cancer sera, arrayed on a glass slide, to probe cancer sera to recognize differences in its proteomic profile compared to normal sera.

Antibodies have the potential to identify biomarkers that are novel, unusually spliced or modified or are present in a differential concentration in cancer serum samples with respect to normal samples. They combine the advantages of an unbiased discovery approach (as is the case for Mass Spectrometry techniques) with the sensitivity of an immunoassay for detecting low abundance serum proteins (such as ELISA). As an added advantage, the antibodies can be used for discovery, purification, identification and characterization of the novel biomarker molecules.

Our techniques allow us to exploit the specific advantages of antibodies as tools to detect differential expression of tumor biomarkers along with the high throughput and low volume requirements of microarray platforms.

Using serum samples obtained from individuals with and without ovarian cancer, we challenged the array to identify antibodies that can correctly predict ovarian cancer status.

Using western blot, immunoprecipitation and mass spectrometry techniques, we have begun to identify and characterize the corresponding antigens which were able to discriminate disease status on our antibody microarray. Our final goal is to elucidate their biological relevance to ovarian cancer and validate their subsequent use as biomarkers for the detection of the disease.

To date, we have run nearly 300 microarray experiments using ovarian cancer serum, ovarian cancer cyst and ascites fluid samples and ovarian cancer cell line lysates. From these experiments some notable results are:

-                Our array platform found that previously published markers for ovarian cancer (CA125, Mesothelin, HE4) could discriminate case and control status and were significantly upregulated in cancer serum.

-                We found additional novel proteins in serum that can discriminate disease status as well (or better) than the previously published markers in two separate arrays with partially distinct sample sets.

-                We validated the microarray results by showing higher levels of the novel proteins in individual cancer serum samples using western blotting.

Other discovery projects we are currently working on are:

• Breast Cancer early detection markers (we have performed over 500 arrays using pre-diagnostic serum)
• Pancreatic Cancer markers using mouse models (in collaboration with Sunil Hingorani's Lab at FHCRC)
• Colon Cancer markers (we have performed over 200 arrays)